What Will Kill Us First: Climate Change or AI?

Or maybe another challenger on the undercard?

There’s a nagging doubt in the back of my mind, whenever I’m warning people about the threats of the climate crisis.

I’m wondering… what if AI gets rid of us all far quicker?!

I don’t base this on any conspiracy theories, I relate my thinking to the many (and increasingly regular) revelations from people who actually worked on the latest AI models — and were so scared at what they saw, they quit and turned whistleblower.

Even the super(ab)users at the top of the AI pile have not been shy about the probability of societal collapse at the instruction of the machines.

AI is now creating AI. Humans are pretty much out of the loop. We don’t really know what’s going on.

So, like I say, it’s a strange one…

I have devoted much of my journalism career to warning about the very real, scientifically-supported dangers of global warming and ecosystem collapse.

And then comes along this bad boy, Artificial Intelligence, and threatens to kill us all before the climate can reach sufficient strength to boil or drown us.

So which one’s going to get us first? AI or climate change?

Or is it something on the undercard? Nuclear war? Economic collapse? Disease?

Take your pick!

Obviously we’re all hoping that humans will wake up sufficiently to fend off every one of these challengers, but we’re both an incredible, and an incredibly stupid species, so I’m not holding out too much hope.

I will cling on to what hope there is, however. To update the words of Muhammed Ali, I’m a fighter, not a doomer.

Just a second of your time please…. Writing on crucially important topics is my full-time job, but I rely on kind donations from readers to continue my work. If you are in a position to support my mission, please consider gifting a donation, however small, at my ko-fi page here:

**https://ko-fi.com/supportmarkcampbell**

Thank you so much…. Mark

Climate and AI: The Direct Link

There are obviously parallels between climate change and AI.

Data centres are spreading like wildfire. And wildfires are spreading like data centres.

The data centres put pressure on local water systems, and devour electricity. When that electricity is provided by oil or gas, that’s creating tons of extra atmosphere-warming emissions.

And, as Emily Atkin expertly reported, the fossil fuel giants are also using the capabilities of AI to increase oil and gas production — contributing to a substantial rise in CO2 pollution.

It’s the fact that AI seemingly has the power to potentially end the human race which bothers me, in more ways than one.

It has been difficult enough persuading people to act on climate change, nature loss and pollution, because so many of the worst-case scenarios have been at least 20 years down the road.

This year has provided extreme weather and associated disasters to such a degree, that far more of us are now waking up to the fact that the climate crisis is here and now.

But it’s not going to have a chance of wiping out humanity within 10 years. And that’s what some AI professionals are actually predicting their technology might do!

I can only hope that people don’t choose an apocalypse to hide from — we need to take action on both climate change, and the rise of the machines.

AI — Where Are We Really?

I’m taking seriously all the revelations that come from former employees of OpenAI, Anthropic and other AI giants. Their experiences appear to be well-founded and well-grounded.

I’m also taking seriously the warnings spoken by the billionaires riding the various AI models in the race to superintelligence. They have been disturbingly honest about AI’s capabilities. They, too, are a little scared that their legacy will be: ‘I helped to wipe out the human race.’

But if nobody is going to be around to debate this, I guess it doesn’t matter too much to them.

It’s important we listen to the whistleblowers. I believe it is only honest of climate communicators to recognise another potential armageddon, whenever it comes along.

So what do those in the know, actually say?

Jacob Coxon

As I write, this is the latest revelation to rock the AI industry. Jacob Coxon is a 27‑year‑old British AI researcher who spent the last three years doing pre‑training work at OpenAI and Anthropic.

On September 9, 2026, he announced on X that he was quitting Anthropic — and the AI industry altogether.

The headlines have centred on his fear that AI systems could “kill us all” by the end of the decade.

Now, that’s got to be treated seriously when it comes from someone who has seen behind the curtain, to how far Artificial Intelligence has progressed without common knowledge.

Coxon wrote, among his lengthy statement:

“Neither company (Anthropic and OpenAI) is acting responsibly.

“They are racing straight to self-improving superintelligence and gambling with our lives.

“These will soon be superhuman systems that can hack anything, revolutionise any field overnight, and acquire real power and resources.

“The people building AI earnestly believe that it could kill us all by the end of the decade. This is not a marketing stunt.

“If anything, many executives and senior researchers will couch their phrasing in the press to sound sensible — but I hear the same people express fear privately.”

Disturbingly, Anthropic’s current, employed, alignment science lead, Evan Hubinger, replied to the post on X and actually supported his former colleague’s main claim.

“Jacob is correct here, we really do earnestly believe AI could kill all humans!

“I personally think it is a more than 10 per cent chance within the next decade.

“I believe Anthropic is trying its best, but we do not yet have a plan to solve alignment for superintelligence and are not clearly on track to.”

WTF??!!

Mr Hubinger later added: “I think the risk from present models is low. What I am worried about is superintelligence arising from recursive self-improvement, as we have said is happening faster than we thought.”

Jacob explains more in this CNN interview with Anderson Cooper:

https://youtu.be/i30jVPqQeOM?si=GoT3OuOK9ZwDTtrb

Let’s take in some other views from the people who actually know what’s going on behind the scenes…

Geoffrey Hinton

The “godfather of AI” believes that there is a 10–20% chance that AI leads to human extinction within about 30 years, via uncontrollable superintelligence or malicious misuse.

Dario Amodei

The Anthropic CEO says AI systems “smarter than Nobel laureates” could arrive by 2027. He also believes that AI-enabled dictatorships are a real possibility.

Demis Hassabis

The CEO of Google DeepMind believes there is a 50% probability of achieving AGI (human-level general intelligence across cognitive functions) by 2030.

Sam Altman

The OpenAI CEO has warned that AI could cause “massive disruption” to white‑collar work, with entry‑level roles hit first, and has acknowledged existential risk from misaligned superintelligence.

Elon Musk

Riding the xAI contender, Musk claims AI will be smarter than any individual human by 2027, and within 10–20 years, work could become “optional” — but we would go through considerable economic pain to get to that point.

Mo Gawdat

The former Chief Business Officer at Google X predicts a “short‑term dystopia” of 12–15 years starting in 2027, with mass job losses, erosion of the middle class, and intensified authoritarian uses of AI.

Gawdat believes that superintelligence could be “a billion times smarter than us by 2045,” and concurs with a 10–20% extinction risk for humanity.

Bill Gates

The Microsoft co‑founder predicts that AI will cause permanent, large‑scale job losses, and that in most occupations, humans will eventually be “vastly inferior to the AI doing the job”.

Yoshua Bengio

The co‑founder of Element AI has been very vocal on the existential risk of AI, and has warned that advanced AI could enable catastrophic misuse through bioweapons.

Open About AI

Where the climate crisis is concerned, you’ll find that the guys who are causing the damage — the oil, gas, and agriculture industry bosses — will defend themselves to the hilt, and dismiss the warnings as nonsense scaremongering.

But the billionaires in charge of AI are not flat-out defending what they are doing. They are entrenched in a race towards superintelligence, and presumably super riches, and freely admit that the technology could cause the end of the human race within a decade.

I can’t believe I just wrote that sentence. I mean, if there’s just a one per cent chance of us all dying, you stop what you’re doing, don’t you? Assess the situation, tweak things a bit…

I do find all this a distracting sideshow when educating people about the climate and biodiversity crises.

Because climate change will definitely make life difficult for humans and all animals in the future. We are already there, actually. It wins on that one, because AI is only just starting to cause problems.

But no mainstream climate body has forecast human extinction.

They forecast catastrophe, famine, displacement, conflict — and these things are already playing out. They will cause great stress and hardship across the planet. People do need to be warned.

But nobody is saying that climate change will cause the end of our species, any time soon.

Artificial Intelligence has that dubious honour sewn up.

Thank you for supporting me here on Medium, and at my ko-fi page here: https://ko-fi.com/supportmarkcampbell